Spatial-temporal dynamics of urban heat island hotspots

Abstract

During the past decades, European cities suffered more heat waves, resulting in more excess mortalities. Although the background climate is getting hotter, this may be amplified by thermal heterogeneity within cities. Urban Heat Island (UHI) effect, which describes the increased temperatures in metropolitan areas compared to its surroundings, is one of the amplifiers of heat waves. In recent years, the improvement of computer performance and the availability of remote sensing data in free and open data archives enable the analysis of long-term, large-scale and high-precision urban land use and land surface temperature (LST) data. While the summer daytime UHI has frequently been studied by thermal remote sensing data, the urban thermal heterogeneity has received less attention. This study aims to characterize the urban thermal heterogeneity of cities by spatial statistical and machine learning method based on multi-source thermal remote sensing data. For this purpose, multivariate parameters of hotspot areas of the UHI are delineated, including land use and land cover (LULC), urban morphology, population density, etc. Based on availability of data, cities of Bavaria serve as the research object for this work. Finally, the spatiotemporal dynamics of the UHI hotpots can be explored.